Enzymes with the ability to cleave N-acetyl neuraminic acid (NANA), also known as sialic acid, from other carbohydrates are present in many microorganisms. These include bacteria such as Vibrio cholerae, Clostridium perfringens, Streptococcus pneumoniae and Arthrobacter sialophilus, and viruses such as influenza virus, parainfluenza virus, mumps virus, Newcastle disease virus and Sendai virus. Most of these viruses are of the orthomyxovirus or paramyxovirus groups, and carry a neuraminidase activity on the surface of the virus particles. Many of these neuraminidase-possessing organisms are major pathogens of man and/or animals, and some, such as influenza virus and Newcastle disease virus, cause diseases of enormous importance.
It has long been thought that inhibitors of neuraminidase might prevent infection by neuraminidase-bearing viruses. Most of the known neuraminidase inhibitors are analogues of neuraminic acid, such as 2-deoxy-2,3-dehydro-N-acetylneuraminic acid (DANA) and some of its derivatives (Meindl et al, Virology, 1974 58 457). Our International Patent Publication No. WO 91/16320 describes a number of analogues of DANA which are active against viral neuraminidase, and it has been shown in particular that 4-guanidino-2-deoxy-2,3-dehydro-N-acetylneuraminic acid (Compound (A), code number GG167) is useful in the treatment of influenza A and B (N. Engl. J. Med., 1997 337 874-880). Other patent applications describe various closely-related sialic acid derivatives (eg. PCT Publications No. WO 95/18800, No. WO 95/20583 and No. WO 98/06712), and anti-viral macromolecular conjugates of GG167 have also been described (International Patent Application No. PCT/AU97/00771).

International Patent Publication No. WO 00/55149, describes dimeric compounds which comprise two neuraminidase binding molecules, such as compound (A), attached to a common spacer or linking group of up to 100 atoms in length.
We have now discovered a novel class of compounds which fall within the generic scope of International Patent Publication No. WO 00/55149, but which are not specifically disclosed therein, and exhibit a surprisingly advantageous anti-influenza activity profile which includes an enhanced lung residency time and high potency.
Without wishing to be bound by theory, the basis for the long residency time in the lungs is thought to be due to the size and molecular weight of the compounds preventing entry through tight junctions in the respiratory epithelium and the polarity of the compounds being such that passage through the cell membranes occurs very inefficiently. An alternative theory is that the compounds themselves interact with the phospholipids in the cell membrane or other components of the respiratory epithelium and increase the residency time in the lungs.